The proposed project is requesting funding to merge and analyze data from the World Health Organization (WHO) Collaborative Study of Alcohol and injuries, currently underway in emergency rooms (ERs) in 11 countries, with data from 33 ER sites, which cover only eight countries, and is presently being analyzed under the Alcohol Research Group's Center Grant. The addition of data from the WHO study to the existing merged data set will bring together the only two such data sets of alcohol and injury in existence (both of which used a similar methodology and questionnaire for data collection), and would double the number of injury cases from 5,274 to 10,774, as well as increase country representation to geographic coverage worldwide. The addition of the WHO data to those presently being analyzed will enhance research aims beyond that which can presently be addressed as follows: 1) the larger number of injury cases will facilitate more fine-grained estimates and allow us to determine the stability of estimates regarding the prevalence of alcohol- related injury and risk of injury and related variables, over time and place, since the 8-country data cover a temporal span of 12 years, and is more limited geographically, while the WHO data will all have been collected within a six-month period and achieves global coverage; 2) these estimates will also be enhanced by restriction of the data to those arriving at the ER within six hours of the injury, which will reduce bias in the recall of drinking-in-the-event and related variables; 3) the analytic potential of inclusion of contextual data (socio- cultural and organizational/administrative variables which were not collected as part of the original WHO study) can be maximized as important explanatory factors in associations of alcohol and injury, across 44 ER sites in 17 individual countries; 4) the opportunity will be provided for new analyses (not possible with the data on hand) of the influence of contextual variables related to a) the validity of clinical assessment of alcohol intoxication, and b) the performance of a brief screening instrument for identifying problem drinking among ER patients. Meta-analytic techniques will be used to examine the consistency of individual-based relationships across ER sites. Causal modeling and hierarchical linear modeling will be used to examine the independent effects and interactive effects of contextual variables with individual-and event-level variables on relationships. Cross-national analyses such as those proposed here will inform the alcohol-injury nexus in the U.S., since the U.S. is composed of many micro-cultures which reflect the socio- cultural and organizational/administrative contextual variables dominating many of the countries in which these data were collected.